using time as an independant varaiable in multip[le regression

[rasuranasinghe]

I have dengue incidences as outcome and as independatnvariables time, rainfall, wind flow. Time is as years. Can I do m ultiple regbrezssion for that

[technocrat]

Ordinary least squares regression with time as an independent variable fails for a variety of reasons. Time series analysis has special tools to deal with this class of problem, in particular. See the {fpp3} suite of packages and accompanying text.

[startz]

I think I have to partially disagree with @technocrat. (That doesn't happen very often!) There's not necessarily anything wrong with time as an independent variable. Indeed, the nice link @technocrat provides gives an example using time as an independent variable.

[technocrat]

Sorry to be obscure @startz. By ordinary least squares regression, I meant lm. Hyndmans text introduces tslm`, a different critter that takes into account the violation of OLS assumptions re normality of residual, independence, etc. it was hard to wrap my head around at first, but autocorrelation , too.

[startz]

Got it @technocrat. I agree that the issues you point out about time series in OLS are very important, but I don't think they have anything to do with whether time is an independent variable.

(And as a pure side issue, for a large sample normality doesn't much matter.)